Vol. I — AI Systems Lab

Designing
Systems
that Think.

I build AI-powered infrastructure for learning, research, and work. Not prompts. Not chatbots. Systems — structured, purposeful, built to help people understand better.

AI Systems Workflow Design Knowledge Architecture
Research Workflow
Knowledge System
AI Agent Loop
Systems Thinking Map — Live
Focus AI Systems · Workflow Design · Automation · Prototyping
Status Independent Consultant · Jan 2023 – Present
Open to Remote · International Teams · EN / RU
AI Systems Design
Workflow Architecture
Knowledge Systems
AI Agents & Automation
Learning System Design
AI-Assisted Prototyping
Research & Experimentation
Prompt Frameworks
AI Systems Design
Workflow Architecture
Knowledge Systems
AI Agents & Automation
Learning System Design
AI-Assisted Prototyping
Research & Experimentation
Prompt Frameworks
01 — Field Notes

Things I believe
about AI systems.

Not conclusions. Evolving convictions — shaped by 3+ years of building, testing, and often getting things wrong first.

01
Systems outperform prompts
A single clever prompt is a trick. A well-designed system is infrastructure. The difference is repeatability, reliability, and the ability to evolve.
02
Understanding matters more than generation
Most AI tools optimise for output volume. The real opportunity is in AI that builds genuine comprehension — in the people who use it, not just the model.
03
Good automation preserves human judgment
The best automation removes friction at the right places — and creates friction at the decisions that actually need a human. Designing that distinction is the hard part.
04
AI should amplify thinking, not replace it
The goal isn't to offload thinking to an LLM. It's to use AI as leverage — so one person can think more clearly, more broadly, and more effectively.
05
Constraints are a design tool
The most useful AI systems aren't the most capable ones — they're the ones with deliberate limits. Knowing when not to respond is as important as knowing how to.
02 —

Domains & Expertise

001
Research Workflow Design
Structuring research processes into repeatable AI-supported systems. From question to synthesis, with documented logic at every step.
002
Knowledge Systems
Building structured architectures that help people understand rather than simply accumulate. Logic, progression, constraint — by design.
003
AI Automation
Intelligent automation flows that eliminate repetitive work while preserving human judgment at every decision point that matters.
004
AI Prototyping
Rapid prototyping of AI-powered tools using modern no-code and LLM-native stacks. From concept to working system, fast.
005
Learning System Design
AI-supported learning experiences focused on understanding, reasoning, and application — not memorisation at scale.
03 —

Experiments & Case Files

All experiments
Experiment 001 · Legal AI · High-Stakes Domain

Belarusian
Legal AI
Assistant

A grounded legal AI assistant built around Belarusian law. Structured retrieval, citation grounding, constrained reasoning — designed with explicit failure modes and trust architecture from the ground up.

RAG Architecture Legal AI Constraint Design Knowledge Grounding

"The challenge wasn't making it answer. It was designing the system to know when not to — and explain why."

Read Case File
Experiment 002 · Education Technology

NotebookLM Learning System

Educational content generation using NotebookLM as core reasoning engine. Transforms dense source material into structured, interactive knowledge — prioritising genuine understanding over output speed.

Case File
Experiment 003 · Content Operations

AI Content Workflow

Reusable prompt workflow infrastructure for content research, drafting, and refinement. Built around structured frameworks — designed to be operated by non-engineers as a repeatable production system.

Case File
Experiment 004 · Claude Skills · Open Source

Claude Skills Library

An evolving collection of reusable Claude skills designed for research, workflow design, learning systems, and practical AI applications. Currently documenting, testing, and publishing skills through open-source experiments and real-world use cases — an active and growing body of work built through craft, not volume.

GitHub Project
View on GitHub
04 —

Research Notes

01
Analysis · AI Economics

Everyone's Generating Videos — I Calculated What AI Video Actually Costs in 2026

A methodical cost analysis of the AI video generation landscape. Separating hype from real economics, comparing major platforms on output quality per dollar — what creators actually need to know before committing budget.

Read Analysis
02
Education · AI Systems

AI for Teachers: How to Create a Lesson Children Will Remember in 2 Hours

A practical framework for educators — a system for using AI in lesson design that prioritises understanding and pedagogical integrity over speed alone.

Read Framework
Essay · In Progress

On Building AI Systems That Serve Understanding — Not Just Output

Why the distinction between AI that produces answers and AI that builds comprehension matters — and what design principles follow from it.

Follow Progress
The best AI systems I've built were defined not by what they could do — but by what they were deliberately designed not to do.
Constraint is a design tool. Knowing when to stop is a feature. Every workflow I design starts with the question: where does human judgment need to stay in the loop — and how do we make that inevitable, not optional?
05 —

Lab Infrastructure

Not a list of logos.
A set of thinking systems — each chosen for a specific role in the lab.
I organize my environment around cognitive roles, not software categories. Every tool earns its place by doing one thing better than anything else in the stack.
Research Stack
Perplexity
Real-time web research
ChatGPT
Broad exploration
Gemini
Multimodal analysis
Knowledge Stack
📓
NotebookLM
Source synthesis
Notion AI
Knowledge management
Custom Prompts
Structured frameworks
Reasoning Stack
Primary
🧠
Claude
Primary system design
AI Agents
Multi-step task execution
Hermes
Agent orchestration
Build Stack
Lovable
AI-first prototyping
Cursor
AI-assisted coding
Vercel
Deploy & ship
Automation Stack
n8n
Workflow orchestration
Zapier
Integration automation
Replit
Rapid app environments
Swipe to explore stacks
06 — Contact · Let's Collaborate

Let's Build
Something
Real.

Open to remote opportunities and international teams. I work best with people who care about depth, clarity, and systems that actually work — not just ship.

Poland · Remote Global · EN / RU
Open to new projects